Application integration has become the 'philosopher's stone' for many indus
trial companies. Many integration projects have succeeded in solving the pr
oblem they set themselves, but left the problem of integration unsolved. Th
e integration of applications for a complex design environment is particula
rly rich in such projects. In the author's experience, a broader view of in
tegration is needed, and several complementary models of integration are de
scribed, together with a list of 12 different design issues that integratio
n projects need to resolve. It is not claimed that a project covering all t
hese features will succeed, but rather an absence of a feature is an indica
tion that the project will fail.